sparknlp.annotator.cv.internvl_for_multimodal
#
Module Contents#
Classes#
InternVLForMultiModal can load InternVL Vision models for visual question answering. |
- class InternVLForMultiModal(classname='com.johnsnowlabs.nlp.annotators.cv.InternVLForMultiModal', java_model=None)[source]#
InternVLForMultiModal can load InternVL Vision models for visual question answering. The model consists of a vision encoder, a text encoder, a text decoder and a model merger. The vision encoder will encode the input image, the text encoder will encode the input text, the model merger will merge the image and text embeddings, and the text decoder will output the answer.
InternVL 2.5 is an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0, maintaining its core model architecture while introducing significant enhancements in training and testing strategies as well as data quality. Key features include: - Large context window support - Multilingual support - Multimodal capabilities handling both text and image inputs - Optimized for deployment with int4 quantization
Pretrained models can be loaded with
pretrained()
of the companion object: >>> visualQA = InternVLForMultiModal.pretrained() … .setInputCols(“image_assembler”) … .setOutputCol(“answer”)The default model is “internvl2_5_1b_int4”, if no name is provided. For available pretrained models, refer to the Models Hub.
Input Annotation types
Output Annotation type
IMAGE
DOCUMENT
- Parameters:
- batchSizeint, optional
Batch size. Larger values allow faster processing but require more memory, by default 1.
- maxSentenceLengthint, optional
Maximum sentence length to process, by default 4096.
Examples
>>> import sparknlp >>> from sparknlp.base import * >>> from sparknlp.annotator import * >>> from pyspark.ml import Pipeline >>> from pyspark.sql.functions import lit >>> image_df = spark.read.format("image").load(path=images_path) >>> test_df = image_df.withColumn( ... "text", ... lit("<|im_start|><image>\nDescribe this image in detail.<|im_end|><|im_start|>assistant\n") ... ) >>> imageAssembler = ImageAssembler() \ ... .setInputCol("image") \ ... .setOutputCol("image_assembler") >>> visualQA = InternVLForMultiModal.pretrained() \ ... .setInputCols("image_assembler") \ ... .setOutputCol("answer") >>> pipeline = Pipeline().setStages([ ... imageAssembler, ... visualQA ... ])
>>> result = pipeline.fit(test_df).transform(test_df) >>> result.select("image_assembler.origin", "answer.result").show(truncate=False)
- setMaxSentenceSize(value)[source]#
Sets Maximum sentence length that the annotator will process, by default 4096. Parameters ———- value : int
Maximum sentence length that the annotator will process
- setIgnoreTokenIds(value)[source]#
A list of token ids which are ignored in the decoder’s output. Parameters ———- value : List[int]
The words to be filtered out
- setMinOutputLength(value)[source]#
Sets minimum length of the sequence to be generated. Parameters ———- value : int
Minimum length of the sequence to be generated
- setMaxOutputLength(value)[source]#
Sets maximum length of output text. Parameters ———- value : int
Maximum length of output text
- setDoSample(value)[source]#
Sets whether or not to use sampling, use greedy decoding otherwise. Parameters ———- value : bool
Whether or not to use sampling; use greedy decoding otherwise
- setTemperature(value)[source]#
Sets the value used to module the next token probabilities. Parameters ———- value : float
The value used to module the next token probabilities
- setTopK(value)[source]#
Sets the number of highest probability vocabulary tokens to keep for top-k-filtering. Parameters ———- value : int
Number of highest probability vocabulary tokens to keep
- setTopP(value)[source]#
Sets the top cumulative probability for vocabulary tokens. If set to float < 1, only the most probable tokens with probabilities that add up to
topP
or higher are kept for generation. Parameters ———- value : floatCumulative probability for vocabulary tokens
- setRepetitionPenalty(value)[source]#
Sets the parameter for repetition penalty. 1.0 means no penalty. Parameters ———- value : float
The repetition penalty
References#
See Ctrl: A Conditional Transformer Language Model For Controllable Generation for more details.
- setNoRepeatNgramSize(value)[source]#
Sets size of n-grams that can only occur once. If set to int > 0, all ngrams of that size can only occur once. Parameters ———- value : int
N-gram size can only occur once
- setBeamSize(value)[source]#
Sets the number of beam size for beam search, by default 1. Parameters ———- value : int
Number of beam size for beam search
- static loadSavedModel(folder, spark_session, use_openvino=False)[source]#
Loads a locally saved model. Parameters ———- folder : str
Folder of the saved model
- spark_sessionpyspark.sql.SparkSession
The current SparkSession
Returns#
- InternVLForMultiModal
The restored model
- static pretrained(name='internvl2_5_1b_int4', lang='en', remote_loc=None)[source]#
Downloads and loads a pretrained model. Parameters ———- name : str, optional
Name of the pretrained model, by default “internvl2_5_1b_int4”
- langstr, optional
Language of the pretrained model, by default “en”
- remote_locstr, optional
Optional remote address of the resource, by default None. Will use Spark NLPs repositories otherwise.
Returns#
- InternVLForMultiModal
The restored model